Current privacy discourse treats data protection as a Boolean state: data is either protected or exposed, consent either given or withheld. This paper argues that the operative variable is decodability — the degree to which aggregated data can be reverse-mapped to an identifiable individual — and proposes a float-valued reconceptualization of privacy governance. Three innovations: (1) a three-stage decodability model decomposing privacy risk into identification, exploitation, and impact, yielding a continuous expected-harm formula E = D1 × D2 × D3 × I; (2) an Otentosama architecture modeled on the Japanese moral concept of an all-seeing but benevolent observer, offering a structural alternative to Foucault's Panopticon; and (3) an integral democracy framework that augments majoritarian voting with continuous optimization of collective welfare through AI computation. The proposed infrastructure combines blockchain (immutable rules), NFTs (portable individual sovereignty), and AI (real-time risk adjudication) in a three-layer separation of powers. The paper positions these proposals against six traditions of data governance — from Zuboff's surveillance capitalism critique to the EU's GDPR framework — identifying their shared Boolean limitation. This is paper 9 of the Boolean-to-Float series (Osada 2026), a ten-paper sequence reformulating discrete categorical judgments as continuous-valued processes across philosophy, ethics, governance, and AI alignment. Version 3.1 incorporates revisions following peer review at three prior venues (E&IT, AI and Ethics, Res Publica) and is co-deposited with submission to Information Polity (SAGE).
Kenshiro Osada (2026) studied this question.